This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [PATCH] PR 32146 (Bad matrix layout transformations)


Diego Novillo <dnovillo@google.com> wrote on 31/05/2007 15:54:29:

> On 5/31/07 8:37 AM, Razya Ladelsky wrote:
> 
> > 2007-05-31  Razya Ladelsky  <razya@il.ibm.com>
> > 
> >         * matrix-reorg.c (transform_access_sites): Fix computation.
> >            (transform_allocation_sites): Same.
> >         *  testsuite/gcc.dg/matrix/matrix-6.c: Remove conversion.
> 
> OK with a couple of nits
> 
> > !          if (x - y == 0)
> 
> This is an odd way to ask 'x == y'.  Any specific reason?

No, I am using the result of their substraction as a shift factor
(in case it is not zero, of course), so this seemed natural to me. 
(I also ask if x-y < 0 or x-y >0)

> 
> > !       dim_size = mi->dimension_size_orig[id];
> > !          else
> > !       {
> 
> No braces needed here.

True, just seemed a little more readable to me.

Thanks for your quick review!
I will make the changes, perform full testing, and then commit.

Razya


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]